DESCRIPTION (Applicant's Description) Forty-five percent of children with neuroblastoma have high-risk disease which is usually metastatic. We recently completed a phase III, randomized study with the Children's Cancer Group that showed myeloablative chemoradiotherapy (supported by purged autologous bone marrow transplantation) followed by 13 cis-retinoic acid improved event-free survival (EFS). However, the estimated EFS from diagnosis still was only 38 percent for patients with metastatic disease. Poor responses or occurrences in primary and metastatic sites, particularly bone and bone marrow, were the causes of failure. Hypothesis - Additional therapies that are not affected by resistance to chemotherapy, irradiation, and retinoic are necessary to further improve treatment. Long-term goal - Develop new therapies based upon understanding the biology of tumor and host cells in metastasis formation, angiogenesis, antibody-mediated neutrophil cytotoxicity, and retinoid (fenretinide) induced tumor cell cytotoxicity. Specific aims - Project 1: To understand the role of matrix degrading proteases and their inhibitors in metastases, emphasizing interactions between neuroblastoma, stromal, and endothelial cells, and to evaluate inhibitors of proteases in preventing and treating metastases. Project 2: To define endothelial cell expression of integrin receptors avB3 and avB5, stromal cell expression of proteases, and proteolyzed collagen in neuroblastoma tumors and to maximize signals from integrin antagonism, especially involving ceramide, that cause apoptosis of endothelial cells. Project 3: To optimize neutrophil cytotoxicity for neuroblastoma cells with an anti-GD2 antibody/GM-CSF fusion protein and with agents that increase neutrophil-tumor cell conjugates and tumor cell sensitivity to neutrophil. Project 4: To develop new therapeutic strategies based upon increasing ceramide induced apoptosis and necrosis with fenretinide and modulators of ceramide metabolism. Project 5: To translate laboratory findings into Phase 1 clinical trials that include pharmacology and dosimetry and assessment of activity within the confines of such studies. Research Design. Projects 1 trough 4 are biologic and developmental therapeutic studies that utilize molecular and cell biology techniques, neuroblastoma tumors from patients, neuroblastoma cell lines growing as primary and metastatic tumors in immunodeficient SCID mice. In Project 5, a consortium of eight institutions will conduct Phase I protocols that are based upon our laboratory studies. Core components will provide research support, pathology, digital image scanning microscopy and flow cytometry, SCID mouse models, and statistics/data management. Conclusion - We anticipate that this program will discover therapies that will significantly improve the outcome of patients with high-risk neuroblastoma.